Subcellular distribution of the glucocorticoid receptor and evidence for its association with microtubules

Subcellular distribution of the glucocorticoid receptor and evidence for its association with microtubules

J. Steroid Biochem. Molec. Biol. Vol. 52, No. 1, pp. 1-16, 1995 Pergamon Elsevier Science Ltd. Printed in Great Britain 0960-0760(94)00155-3 Revie...

2MB Sizes 10 Downloads 88 Views

J. Steroid Biochem. Molec. Biol. Vol. 52, No. 1, pp. 1-16, 1995

Pergamon

Elsevier Science Ltd. Printed in Great Britain

0960-0760(94)00155-3

Review S u b c e l l u l a r D i s t r i b u t i o n of the G l u c o c o r t i c o i d R e c e p t o r and E v i d e n c e for its Association with Microtubules o

Gunnar Akner, A n n - C h a r l o t t e Wikstr6m and Jan-Ake Gustafsson* Department of Medical Nutrition, Karolinska Institute, Huddinge University Hospital, Novum F60, S-141 86 Huddinge, Sweden The cellular distribution of the glucocorticoid receptor (GR) has not yet been firmly established. The e x t e n s i v e l i t e r a t u r e i n d i c a t e s t h a t G R is p r e s e n t b o t h in t h e c y t o p l a s m a n d t h e cell n u c l e u s , h o w e v e r , s o m e s t u d i e s h a v e f a i l e d to d e t e c t c y t o p l a s m i c GR. It is still c o n t r o v e r s i a l as to w h e t h e r G R is r a n d o m l y d i f f u s i n g in t h e c y t o p l a s m a n d n u c l e u s , o r i f t h e G R - d i s t r i b u t i o n is o r g a n i z e d o r c o n t r o l l e d in s o m e w a y , w h i c h m a y b e o f i m p o r t a n c e f o r t h e t r a n s d u c t i o n o f g l u c o c o r t i c o i d effects to cells. T h e r e is e v i d e n c e t h a t b o t h n o n - a c t i v a t e d a n d a c t i v a t e d G R is a s s o c i a t e d w i t h t h e p l a s m a m e m brane, a number of cytoplasmic organelles and the nucleus. Both morphological and biochemical e v i d e n c e s h o w t h a t G R is a s s o c i a t e d w i t h m i c r o t u b u l e s d u r i n g d i f f e r e n t s t a g e s o f t h e cell cycle, i.e. G R c o - l o c a l i z e s , c o - p u r i f i e s a n d c o - p o l y m e r i z e s w i t h t u b u l i n . T h i s i n d i c a t e s t h a t G R is s t r u c t u r a l l y l i n k e d to t h e i n t r a c e l l u l a r M T - n e t w o r k w h i c h m a y b e o f i m p o r t a n c e in t h e m e c h a n i s m o f a c t i o n o f g l u c o c o r t i c o i d h o r m o n e s . T h e l i t e r a t u r e in t h i s field is r e v i e w e d i n c l u d i n g t h e r e p o r t e d d a t a o n subcellular GR-loca][ization.

J. Steroid Biochem. Molec. Biol., Vol. 52, No. 1, pp. 1-16, 1995

BIOCHEMICAL CHARACTERISTICS OF GLUCOCORTICOID R E C E P T O R

THE

T h e glucocorticoid receptor (GR) is a ligand-activated transcription factor be]onging to the steroid hormone receptor superfamily. Protease degradation of G R defined three distinct, functional domains [1, 2] which were later confirmed at the G R c D N A level using site-directed mutagenesis [3]. T h e transactivating (N-terminal, immunogenic) domain is the least conselwed domain among the various members of the steroid receptor superfamily. Most mono- and polyclonal a.ntibodies against G R recognize this domain. T h e D N A - b i n d i n g domain is highly conserved in all steroid receptors. T h e core region of this domain contains two zinc fingers, analogous to the zinc-finger regions of T F I I I A and other transcription factors responsible for sequence specific D N A - b i n d i n g [4, 5]. T h e hormone binding (C-terminal) domain is *Correspondence to J.-A. Gustafsson. Received 5 July 1994; accepted 5 Sep. 1994.

required for high affinity steroid binding [6] and there is also evidence that this domain binds the hsp90 dimer [7]. Deletions within this domain of G R results in a protein with constitutive transactivating capacity [8]. T h e hormone-binding and D N A - b i n d i n g domains are separated by a " h i n g e " region, containing a short stretch of highly basic amino acids [9]. This region has been implicated in nuclear localization of the rat G R [10] and is homologous to sequences required for nuclear localization of other proteins [11]. Further analysis of different parts of G R by deletion mutagenesis has revealed other functional regions as well. T h e r e are e.g. two separate transactivation regions within GR, localized in the N-terminal domain (z 1) and between the D N A - and steroid binding domains (z2), respectively [12]. T h e amino acid sequence of the human GR, deduced from sequence analysis of c D N A clones, revealed the existence of two major isoforms of GR, ~ and /3, 777 and 742 amino acids in length, respectively, generated by alternative splicing of an m R N A encoded from a single gene on human chromosome 5 [12]. In the rat G R (795 amino acids), the three domains contain

2

Gunnar Akner et al.

409, 108 and 278 amino acids, corresponding to ~ 5 1 , 14 and 35 ~o of the primary sequence, respectively [13]. During basal conditions, the non-activated G R is present in the cell as a multimolecular, heterooligomeric complex together with several other proteins, e.g. a dimer of the heat shock protein MR90,000, hsp90 [14], the immunophilin hsp56 [15, 16] and p23 [17]. T h e r e is also evidence that the complex may contain other components, e.g. R N A [18] and phosphate [19]. Regarding tubulin, see below. T r e a t m e n t of cells with glucocorticoid hormones in the presence of heat causes activation of GR, i.e. the receptor acquires DNA-binding ability [20]. Thus, at +4°C, the hormone only binds to, but does not activate GR. It is unclear why heat ( + 20°C, 30 min) is needed along with the ligand. This heat-requirement for activation represents an in vitro phenomenon and can not be of physiological significance in mammals in vivo. Some features of the activated G R are presented below. Features of the activated human GR: --Macromolecule, MR94,000 [21] - - 7 7 7 amino acids [12] - - S e d i m e n t a t i o n coefficient ~ 4S [21 ] - - A s y m m e t r i c protein, with a length of approx. 0.12-0.15 nm [22] - - N e g a t i v e net charge [19] --Acidic protein, pI 5.7 [23] - - P h o s p h o p r o t e i n [19] - - O n e molecule of glucocorticoid aporeceptor binds one molecule of glucocorticoid hormone [24] - - T h e activated G R is able to bind both to nonspecific D N A and glucocorticoid response elements (GREs) [2].

TISSUE LOCALIZATION OF GR G R has been detected in many different mammalian tissues, however, there are reports that certain tissues (rat) are devoid of GR, e.g. the intermediate lobe of the pituitary [25], liver Kupffer cells and liver endothelial cells [26], uterus, prostate gland, seminal vesicles, bladder, adipose tissue and jejunum [27], kidney glomeruli and proximal convoluted tubules [28] and acinary cells in submaxillary glands [29]. It has also been claimed that neither rat neurons nor rat lymphocytes contain G R [30], but this is in contrast to other reports [31, 32]. MODELS OF GLUCOCORTICOID H O R M O N E LOCALIZATION AND ACTION T h r e e commonly discussed models of steroid hormone mechanism of action (Fig. 1) have all been applied to GR. Early studies on steroid hormone receptor localization focused on the estrogen receptor (ER) and were based on cell fractionation experiments

ACTIVATION MODELS

a. Two-step model

Rda

b. Equilibriummodel

c. One-step model Fig. 1. Activation models. Three c o m m o n , hypothetical m o d e l s of steroid h o r m o n e action. R, receptor; GRE, glucocorticoid response element; p, phosphate; c, cytoplasm; n, nuclear; s, soluble; b, bound. The glucocorticoid h o r m o n e is represented by a grey triangle containing an "s" for steroid. See also text.

supported morphologically be cellular autoradiography [33]. These studies led to the development of "the two-step m o d e l " of estrogen hormone action [33] and later to a unified theory encompassing all steroid hormone receptors including G R [34] [see Fig. l(a)]. T h e model stated that the non-liganded steroid hormone receptor is soluble in the cytoplasm/cytosol of the cell. After binding of a specific steroid ligand, the receptor complex undergoes a temperature-sensitive process denoted "activation" or "transformation", during which the receptor complex acquires an increased affinity for chromatin or D N A to alter gene expression. However, the model was soon challenged [35]. New techniques for autoradiographic sample processing [36] lead to new conclusions and also to reinterpretation of some of the previous autoradiographic data. T h e twostep model was thus replaced by an "equilibrium

Subcellular Distribution of Glucocorticoid Receptor Table 1. Classification of non-liganded steroid hormone receptors according to their affinity for nuclear components (Adapted from Sanchez et al., 1990142]) Group 1 Tightly bound to nuclei even in the absence of ligand. High salt required for extraction from nuclei, e.g. Thyroid hormone receptor, retinoic acid receptor Group 2 Relatively weakly bound to nuclei in the absence of ligand. Recovered in cytosolic fraction after lysis of cells in hypotonic buffer. Become tightly associated to nuclei after treatment with ligand and high salt is then required for extraction, e.g. Estrogen receptor, progesterone receptor, androgen receptor (?) Group 3 Located in the cytoplasm in the absence of ligand. Recovered in cyto,;olic fraction after lysis of cells in hypotonic buffer. Become tightly associated to nuclei after treatment with ligand and high salt is then required for extracxion, e.g. GR, 1,25-dihydroxy-vitamin D3-receptor (?)

m o d e l " , originally for the E R and progesterone receptors (PR) [36, 37] and then expanded to all steroid receptors [38] [see Fig. l(b)]. T h i s model postulated an equilibrium between the cell b o u n d and soluble forms of steroid receptors both in the nucleus (bound to chromatin) and the cy~:oplasm (bound to ?) based on partitioning according to the free water present in each c o m p a r t m e n t , fixed charges, p H , ionic strength etc. T h e s e parameters were expected to vary between tissues and species. Available data was interpreted such that for both E R and PR, at least part of the u n b o u n d receptor resided in the nucleus, and that it was extracted into the cytosolic fraction during homogenization and separation [[38]. Recent studies using P R deletion mutants indicate that the nuclear localization of PR is due to such a dynamic equilibrium: PR diffusing into the cytoplasm is constantly and actively being transported back into the nucleus [39]. T h i s nuclear-cytoplasmic shuttle energy-dependent mechanism is not dependent on an intact cytoskeleton [40]. Immunolocalization experiments have provided evidence that several of the m e m b e r s in the steroid h o r m o n e rceptor superfamily are localized in the cell nucleus at all times [41]. T h i s led to the "one step m o d e l " , implying that both non-liganded and liganded steroid h o r m o n e receptors are nuclear proteins and that the lipophilic ligands traverse both the plasma m e m brane and the nuclear envelope and bind to their respective receptors in the nucleus directly [see Fig. l(c)]. T h e older biochemical data have then been reinterpreted and the cytosolic localization of receptors in cell extracts is claimed to be due to artifactual redistribution of receptors during cell- or tissue-horn-

3

ogenization. Attempts have even been made to group the steroid receptors according to their various degree of extractability [42] (see T a b l e 1 for a summary). NUCLEAR T R A N S L O C A T I O N T h e concept of ligand induced nuclear translocation of G R has been widely accepted, even though several studies have failed to demonstrate such a process [43-47]. In any case, it seems clear that a part of the total G R - p o p u l a t i o n m u s t undergo at least one nuclear translocation, i.e. after being synthesized in the endoplasmic reticulum in the cytoplasm. Macromolecules of the size of G R probably enter the nucleus in a controlled fashion [48] and require specific signal sequences for nuclear uptake [49]. By analogy to mitochondrial and endoplasmic proteins, controlled passage of large proteins across the nuclear envelope should require (i) a signal for nuclear migration within the protein itself and (ii) a mechanism at the nucleus to respond to the signal. T h e r e is evidence suggesting that nuclear envelope proteins, with MR60,000 and 76,000, respectively, interact with nuclear localization signals of G R [50]. By experiments using deletion mutants and fusion proteins, Picard and Y a m a m o t o found that G R contains two independent nuclear localization signals [10]. T h e first signal, N L 1 , is 50% homologous to the SV40 large T antigen nuclear localization sequence and located just at the C-terminal side of the D N A - b i n d i n g domain (the "hinge"-region). T h e signal is functionally repressed when the steroid binding domain is present, but becomes constitutively active when this G R - d o m a i n is truncated. Nearly identical sequences are found in G R , P R and in androgen and mineralocorticoid receptors. In contrast, sequences in this region of the ER and 1,25-dihydroxy-vitamin D3 and thyroid h o r m o n e receptors do not exhibit strong homology to the T - a n t i g e n N L S [51]. T h e second signal, N L 2 , lies within the steroid binding domain and has not been separated from hormonal control. T h e nuclear translocation of G R is reported to be fast with a T~/2 of 1-5 min at + 3 7 ° C [10]. GR-RECYCLING T h e fate of G R after having exerted its gene regulatory effect(s) is unclear. Theoretically, G R may be degraded or reutilized. T1/2 of the G R - m R N A is 4.5 h and is unaffected by hormone [52]. Tx/2 of the G R -

Table 2. Classification and composition of the three main cytoskeletal networks (Adapted from Alberts et al., 1989156]) CytoskelLetal network Microtubules (MTs) Intermediate filaments (IFs) Microfilaments (MFs)

Polymer subunit Tubulin (~- and fl-isoforms) Severaldifferent IF-proteins Actin

O

MR

25 nm 8-11 nm 5-7 nm

55,000 40,000-210,000 42,000

4

Gunnar Akner et al.

protein in the absence of hormone is 20-25 h, in the presence of hormone 9-11 h [52,53]. Glucocorticoid hormones have dissociation half-times off G R of 2 m i n (cortisol, corticosterone) and 10-30 min (dexamethasone, triamcinolone acetonide), respectively [54]. T h e r e is evidence supporting a recycling mechanism of GR: activated nuclear G R is recycled back to the cytoplasm where it is deactivated [54, 55] followed by a net synthesis of G R [55]. Evidence has also been presented that e.g. PR is recycling between the nucleus and cytoplasm (see above). I N T E R A C T I O N B E T W E E N GR A N D THE

CYTOSKELETON T h e r e are three main cytoskeletal networks in mammalian cells, divided according to the diameter (O) of the skeletal "fibers" [56]. T h e y are all polymers of different protein subunits (see Table 2). Several observations suggest that G R may be linked to the microtubule ( M T ) part of the cytoskeleton (see below) and there is evidence that centrioles can bind steroid hormones such as 17flestradiol, progesterone and testosterone [57]. T h e r e are also reports that glucocorticoids and G R may act through the microfilament (actin) system. G R binds to actin filaments through hsp90 [58] and treatment of cells with glucocorticoids stabilizes actin networks [59,60]. It is thus conceivable that glucocorticoid hormones transduce some effect(s) to cells through an interaction between G R and the intracellular cytoskeleton. CHARACTERISTICS OF MICROTUBULES Microtubules (MTs) are ubiquitously distributed throughout the animal and plant kingdoms. T h e y are hollow, unbranched, tubular cellular organelles of variable length and with an outer diameter of about 25 nm and are found in all nucleated eukaryotic, but not procaryotic, cells [61, 62]. T h e M T - c y l i n d e r contains a central core with unknown composition or function [62]. T h e walls of cytoplasmic M T s are composed of 13 subunits (protofilaments), which are aligned parallel to the long axis and folded into a cylinder. Each protofilament is built up by polymerized heterodimers of a- and fl-tubulin. T h e M T - s t r u c t u r e is similar in cytoplasmic M T s , mitotic apparatus M T s , centriolar/basal body M T s and ciliar/flagellar M T s . In interphase cells, associations have been observed between cytoplasmic M T s and most intracellular organelles and membranes [62]. Such interactions may be direct links between the tubulin polymer and the respective organelle or indirect links via microtubule associated proteins (MAPs, see below).

CHARACTERISTICS OF TUBULIN T h e basic subunit of M T s is tubulin, a 6S heterodimer of two globular polypeptide chains (a- and fl-tubulin) with an MR of 100kDa, each monomer having an MR of 50 kDa [63]. T u b u l i n exhibits very strong species homology through evolution: antibodies against sea urchin egg tubulin cross-react with tubulin from most species and tissues, including humans [62]. Both human ~- and fl-tubulin have been cloned and deduced to consist of 452 and 444 amino acids, respectively. In humans, each of the genes encoding a- and fl-tubulins, constitutes a large multigene family of about 15-20 members [63, 64] only a portion of which represents functional isoforms. T h e a- and fl-tubulins exhibit around 50% D N A sequence homology, suggesting a common ancestral gene. T u b u l i n constitutes one of the most abundant cellular proteins and comprises 2 - 3 % of total cellular protein. Only 40% of the intracellular tubulin pool is polymerized during interphase [65]. It is not known how the pool of soluble tubulin ~-fl-dimers is distributed throughout the cell and/or whether it is compartmentalized. T h e M T - p o l y m e r contains a number of distinct binding sites, e.g. for various MAPs (MAP1B, MAP-2, tau), nucleotides ( G T P , cAMP), Ca 2+ and drugs (colchicine, vinblastine, actinomycin D, taxol, neuroleptic drugs) [66]. MT-FORMATION In the cell, M T - a s s e m b l y is unidirectional originating from discrete foci known as M T O C s (microtubule organizing center), where M T s are nucleated proceeding towards the plasma membrane [67]. Mammalian cells contain several M T O C s , e.g. the centrosome, serving as a nucleation site for the interphase cytoplasmic M T s , and kinetochores, which nucleate M T s during cell division. T u b u l i n polymerization occurs from one or several organizing centers towards the cell periphery. Drug-induced net depolymerization, on the other hand, starts in the periphery and progresses inwards. Exposure of cells to temperatures < + 10°C depolymerizes most cytoplasmic M T s in less than 30 min [62]. CHARACTERISTICS OF M I C R O T U B U L E ASSOCIATED PROTEINS (MAPS)

MAPs may be classified into structural MAPs (e.g. M A P - l , MAP-2, tau, hsp70), translocator MAPs (e.g. kinesin, dynein) and other MAPs (e.g. various enzyme activities, calmodulin, ankyrin) [66, 68-71]. T h e following criteria have been used to define MAPs [72]: (1) they induce assembly of purified tubulin in vitro in the presence of G T P and Mg 2+, (2) they co-polymerize with purified tubulin during repeated cycles of temperature-induced polymerization/

Subcellular Distribution of Glucocorticoid Receptor depolymerization maintaining a constant stoichiometric relationship with tubulin; and (3) they co-localize with cellular M T s . T h e assembly-induction is now believed to constitute an in vitro p h e n o m e n o n for m a n y M A P s and is thus not an obligatory criterion. M o r e emphasis is focused on structural interaction with tubulin. T h e t e r m " M T - b i n d i n g proteins" would be a more accurate general designation, considering that some of these proteins seem to interact only transiently with M T [66]. A S S O C I A T I O N B E I ' W E E N MAPS AND M T S Structural M A P s and translocator M A P s interact with different parts of the tubulin molecule [71, 73]. M A P - 2 and the tau-proteins bind to the M T - p o l y m e r through a cationic M T - b i n d i n g m o t i f characterized by 3-4 highly conserved repeats of 18 amino acids [74]. M A P - 1 has a different repeating m o t i f responsible for M T - b i n d i n g [69]. T h e r e is often a characteristic molar ratio for M T s and M A P s , suggesting a fixed spacing between the M A P s and the tubulin heterodimers along the M T backbone. T h i s is in line with ultrastructural observations indicating a regular interval between various M T - e x t e n s i o n s . T h e average periodicity for both M A P - 2 and tau is around 100 n m [75] corresponding to one M A P - 2 per 14 tubulin dimers and one tau per 17 tubulin dimers [75]. In the living cell, M T s are probably totally saturated with M A P s [76]. Although each tubulin m o n o m e r has a capacity to bind e.g. M A P - 2 and tau-proteins, the actual interaction is more infrequent. T h i s may be due to e.g. steric hindrance. ASSOCIATION BETWEEN TUBULIN/MAPS AND DNA Purified tubulin alone does not bind to D N A , however, tubulin in the presence of M A P s as well as M A P s alone strongly bind to D N A in vitro [77]. T h e tubulin/MAP mixture binds preferentially to satellite D N A - s e q u e n c e s in the eukaryotic genome. Such sequences are usually not transcribed and are located in the chromatin associated with the centromeric regions of the chromosomes. C H A R A C T E R I S T I C S OF T H E M I T O T I C SPINDLE T h e mitotic apparatus is composed of M T s organized as a bipolar spindle. T h e r e is biochemical and morphological evidence that the mitotic spindle, besides tubulin, consists of a n u m b e r of different proteins, such as various NIAPs (MAP-1 [78], tau [79], kinesin [80], dynein [81], heat shock proteins (hsp90 [82], hsp70 [83]), ankyrin [68], calmodulin [84], various enzymes [85-87], myosin [88], actin [89], proteasomes [90] and G R [91, 92]. Some of these components are also associated with the cytoplasmic M T s .

5

C H A R A C T E R I S T I C S OF C E N T R I O L E S Centrioles have been observed in three distinct locations within the cell: (i) in the interphase centrosome; (ii) at the mitotic spindle poles; and (iii) in basal bodies just beneath cilia and flagella. T h e centrosome of interphase animal cells is localized above or at the edge of the nucleus next to the nuclear envelope [93] and typically consists of a pair of orthogonally arranged centrioles surrounded by an osmiophilic matter called the pericentriolar material comprising the centrosomal M T O C . In some cells, the pericentriolar material is distributed primarily around the older of the two centrioles. T h e typical centriole is a small, hollow cylinder, open at both ends unless it is ciliated [94]. T h e centriole is mostly composed of the M T - p r o t e i n tubulin, but also other proteins, e.g. M A P - l , calmodulin and actin [95]. F u r t h e r m o r e , various enzyme activities have been found to be associated with centrioles, i.e. a protein kinase [96] and mechanochemical A T P a s e s such as kinesin [80] and dynein [81]. Some reports indicate the presence of nucleic acids [94, 97], however, no direct evidence is available and this issue is highly controversial. S U B C E L L U L A R L O C A L I Z A T I O N OF GR T h r e e different types of methods have c o m m o n l y been used to localize G R in cells.

Cell fractionation Intact cells, with or without treatment with radiolabelled hormones, are ruptured by any of a n u m b e r of different procedures, e.g. piston-homogenization, ultrasonic sound (sonication), shearing (e.g. by a Polytrone ®) or detergent-induced cell lysis. T h e h o m o g e n ate is then ultracentrifuged at ~ 100,000g to obtain a cytosolic and a nuclear preparation, sometimes with intermediate centrifugation steps. T h e various biochemical fractions are analyzed, e.g. by detection of a radiolabel or by immunochemistry. A major drawback of all assays performed on tissue extracts lies in their inability to provide information about inter- and intra-cellular distributions of a certain component. It is often unclear whether various biochemical fractions really represent specific cellular c o m p a r t m e n t s in vivo, or if there is leakage of water

Table 3. Summary of previous results of immunolocalization of GR including effect of ligand GR in both cytoplasm and nucleus, in various proportions Effect of added ligand: 1. No effect [43, 44] 2. Partial translocation of GR to the nucleus [106, 109, 111] 3. Complete translocation of GR to the nucleus [10] GR only in the nucleus [45, 46] Effect of added ligand: No effect of ligand has been reported in studies that claim solely nuclear GR-localization

6

Gunnar Akner et al.

and/or molecules between the compartments during sample preparation. Even though the interaction between ligand-receptor on the one hand and receptor-acceptor sites on the other is strong, it is non-covalent and thus, both the steroid and its receptor are subject to possible displacement during experimental manipulation of receptor preparations. T h e "cytosol" has unfortunately often been referred to or thought of as representing the "cytoplasm". Similarly, pellet fractions have often been referred to as "nuclear fractions", without realizing that several components of the cytoplasm will probably always accompany nuclear preparations: the rough endoplasmic reticulum is e.g. continuous with the outer nuclear lamina [56], and several components of the cytoskeleton, i.e. microtubules, are also closely associated with the nucleus [62]. It has been suggested that lysosomal macromolecules, e.g. estrogen receptor (ER) like proteins, may be extruded from fragile lysosomes during vigorous homogenization in hypoosmotic media and thus may contaminate the "cytosol" supposed to contain only "soluble" proteins [98]. Regarding ER it has been claimed that the amount of apparent nuclear receptor varies greatly depending upon the procedure used to prepare the nuclear and cytosolic fractions [37]. However, many such investigations of GR-distribution have been interpreted as providing support for G R operating through "two-step model of steroid hormone action" (see below). Another fractionation technique is enucleation, accomplished by centrifugation of intact cells in the presence or absence of cytochalasin B followed by detection of the receptor in the resulting nucleoplasts and cytoplasts [99]. However, here also, cytoplasmic structures connected to the nucleus may contaminate the "nuclear" fraction. Cellular autoradiography

Intact cells are incubated with radiolabelled hormone. After appropriate washes, the cells are exposed to a photographic film and examined by the microscope. T h e r e are several recognized problems with this localization technique: (i) the added hormones may alter the distribution of receptor molecules after binding to their receptors; (ii) there may be redistribution of hormones during sample preparation for autoradiography; and (iii) long film exposure times are usually required (months to years) in order to obtain strong enough 3H-signals for visualization. Studies using 3Hcortisol, 3H-dihydrotestosterone and 3H-17/~-estradiol have shown that there is both nuclear and cytoplasmic localization of 3H-cortisol [100, 101], but mainly nuclear distribution of the receptors for the sex steroids [38, 102]. Immunocytochemistry

This term encompasses immunological localization procedures using antibody-based detection techniques.

Both immunocytological and immunohistological studies are included under this heading. Fixed cells are labelled with specific mono- or polyclonal antibodies and detected by labelled secondary antibodies, representing an indirect detection. Direct immuno-detection implies the use of labelled primary antibodies. M o n o clonal antibodies against G R became available during the mid 1980s [103-105]. This localization technique has several advantages. (i) High specificity, especially when using several different monoclonal antibodies. (ii) Antibody-based detection constitutes a ligand-independent detection system. This eliminates the problem of putative hormone-induced redistribution of receptors and receptors may be detected, even at low concentrations, in tissues taken from animals or patients with high endogenous hormone concentrations. Furthermore, this method identifies the total immunologically reactive G R present, while hormone-binding techniques only label non-liganded GR. (iii) T h e technique is rapid, reproducible and allows the detection of several components in the same cells (double or triple staining) by using specific primary and secondary antibodies with different detection systems. T h e major disadvantage is that the cells have to be fixed and permeabilized to allow access of the antibodies to the inner parts of the cells. T h e r e may be significant intracellular redistribution of substances during sample processing before immunostaining. One way to circumvent this inherent problem is to combine several different fixation/permeabilization techniques using different chemical principles. Taken together, this technique has become the method of choice for localization of cellular proteins. A large number of studies regarding the localization of G R in cells or tissues have been presented. T h e conflicting evidence regarding immunolocalization of G R are summarized in Table 3. In many previous immunological studies, G R has been reported to be diffusely distributed in the cytoplasm [43, 44, 106]. G R has, however, also been reported to interact with subcellular organelles or proteins, based both on biochemical and morphological techniques, see Table 4. T h e r e is evidence that both non-activated and activated G R binds specifically, strongly and preferentially to histones H3 and H4 and this histone-bound G R may thus represent nonextractable forms of nuclear G R [107, 108]. T h e Table 4. Evidence of interactions between GR and cell organelles. The evidence is both biochemical (b) and morphological (m) in nature

Plasma membrane: m, b [32, 135] Microfilaments: b [58] Microtubles: m [91, 92, 111], b [ 112-114] Ribosomes, both free and membrane bound: m [136] Endoplasmic reticulum: m [136] Mitochondria: m [137], b [138] Nuclear envelope: m [139], b [140]

Subcellular Distribution of Glucocorticoid Receptor functional significance of this interaction is unknown, but may be related to the changes in nucleosome conformation that occur during transcription [107, 109]. It is noteworthy that tubulin has also been shown to both localize in intranuclear spots [83] and to interact with histone proteins with a similar predilection for different histones as G R [110]. I N T R A C E L L U L A R O R G A N I Z A T I O N OF GR As outlined above, there is a confusion in the literature regarding the subcellular distribution of G R in cells in tissues and in culture. We consider it unlikely that a macromolecule like the G R aporeceptor would be freely diffusing in the cytoplasm; it would rather be connected to some intracellular structure(s) in order for the cell to transduce and regulate the receptor mediated function(s) adequately. T h e r e are several indications in support of such an organized intracellular d.istribution of GR: (i) i m m u n o cytological studies have shown that G R colocalizes with tubulin during the whole cell cycle in cultured m a m malian cells, with or wi~:hout treatment with glucocorticoid hormones [91,912, 111]; (ii) tubulin copurifies with liganded G R from rat liver [91]; (iii) activated G R in L-cell cytosol is converted from soluble to particulate f o r m under conditions that favor M T polymerization [112-114]. T h e particulate material contains several cytoskeletal components including tubulin, actin and vimentin. T h e C-terminal half of the receptor was necessary and sufficient for this association of G R with the cytoskeletal complex and D N A - b i n d i n g activity was not required [114]. Recently, it has been reported that another m e m b e r of this receptor family, the vitamin D receptor, interacts transiently with M T s during a few minutes after ligand binding [115]. C O M M E N T S ON P R E V I O U S L I T E R A T U R E REPORTS T h e r e are several possible explanations for the conflicting literature reports regarding GR-distribution.

7

Methodological Results from immunocytological and immunohistological studies, as well as results from various fixation/permeabilization techniques used in either cytological or histological studies have been c o m p a r e d directly, as if the various ways of processing a tissueor cell-sample before localization analysis might not affect the apparent subcellular localization of GR. On the contrary, various processing techniques m a y influence both the actual intracellular distribution of antigens, as well as allowing detection of certain, but not other, antigens, due to variations in accessibility of antigens and/or antibodies.

Biological It is conceivable that different cell- and tissue-types may present differences regarding the precise intracellular distribution and function of G R , depending on differences in e.g. germinal layer origin, developmental stage, degree of cellular differentiation and phase of the cell cycle. Morphological experiments constitute a necessary c o m p l e m e n t to biochemical studies for providing an understanding of the function of steroid receptors.

C O M M E N T S ON M E T H O D O L O G Y IN OUR OWN LOCALIZATION STUDIES

Cell Types We have studied a n u m b e r of different cell types, both primary cultures and established cell lines, representing different germinal layers and m a m m a l i a n species/organs [47]. We chose to concentrate the studies mainly on fibroblasts because these cells: (i) are target cells for endogenous and exogenous glucocorticoid action; (ii) are large cells which make them well suited for subcellular localization studies; (iii) exhibit a characteristic and stable phenotype, easily visually determined in the microscope; and (iv) require relatively simple culture conditions.

Table 5. Subcellular distribution of GR in cultured cells Conceptual T h e biochemically defined "cytosolic" G R (i.e. water soluble supernatant after high speed centrifugation) on the one hand and the morphologically defined " c y t o p l a s m i c " G R on the other have been c o m p a r e d as if "cytosol" and " c y t o p l a s m " were synonymous. T h e fact that G R is recovered in the cytosol in vitro does not necessarily mean that G R is a freely diffusing, water soluble macromolecule in vivo. F u r thermore, the observed h o r m o n e induced change in G R - d i s t r i b u t i o n in vitro, i.e. cytosolic in the absence and b o u n d to the "nuclear pellet" in the presence of glucocorticoids, is not necessarily equivalent to a horm o n e induced c o m p a r t m e n t shift in vivo.

General distribution Cytoplasmic pattern Centrosome (MTOC) Granular pattern along fibrils Nuclear pattern Nuclear envelope Plasma membrane (intermittent) Vesicle membrane Mitotic apparatus (all mitotic stages) Cellular protrusions induced by

n+c fibrillar or diffuse ++ + diffuse and granular (+ ) + + + +, especially centriolar regions

treatment with

MT-depolymerizing drugs Vinblastine induced paracrystals

+ +

+, present; n, nucleus; c, cytoplasm; MTOC, microtubular organizing center [47, 91].

Gunnar Akner et al.

ill

F-----4

Fig. 2. Double s t a i n i n g of GR a n d t u b u l i n . A, B, C, D: C L S M - p r o d u c e d t r a n s v e r s a l , 1 g m t h i n optical sections t h r o u g h two well s p r e a d h u m a n gingival f i b r o b l a s t s (A/B a n d C/D) d o u b l e s t a i n e d for G R (A, C) a n d t u b u l i n (B, D) using specific m o n o c l o n a l a n t i b o d i e s a n d i n d i r e c t i m m u n o f l u o r e s c e n c e detection. G R is d i s t r i b u t e d b o t h in the n u c l e u s a n d t h e c y t o p l a s m . T h e r e is m u c h m o r e G R t h a n t u b u l i n in t h e nucleus. C y t o p l a s m i c GR is colocalized w i t h m i c r o t u b u l e s . G R is d i s t r i b u t e d w i t h i n t h e m i t o t i c spindle a p p a r a t u s (C; m e t a p h a s e ) , w h e r e it is colocalized w i t h tubulin (D). Bar corresponds to 20 ~ m (A, B) and 40 t~m (C, D).

In order to achieve as " n o r m a l " a situation as possible, we focused on human, primary culture fibroblasts derived from explants of gingival mucosa. These cells exhibit a typical morphological fibroblast phenotype in the whole cell population in a monolayer, they remain stable for at least 30 subcultures and are generally easy to handle during cell culture and splitting. Fixation/Permeabilization

We tried a n u m b e r of different fixation/permeabilization techniques with principally different chemical

mechanisms (crosslinking or precipitating fixation). We focused on a comparison between two techniques. (1) Cross-linking fixation: typically, we used 4% formaldehyde at + 4 ° C , p H 7.4 for 10min followed by 0.1% (v/v) T r i t o n X-100 ~ for 30 min, Formaldehyde at this concentration, p H and incubation time allows rapid crosslinking. At 5 % (w/v) concentration, formaldehyde reacts preferentially with the e-amino groups of the lysines, forming polymethylether crosslinks with imino-acetals at their reactive sites [ 116]. (2) Precipitating fixation: typically we used methanol at - 2 0 ° C for 10 min.

Subcellular Distribution of Glucocorticoid Receptor

9

~:~:~:~:~:~::~:~:~:~:~~::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: : ~: : i i i: i: i: i: i i: i: i i i i i~i i i~i~i: i ~:;:~i~: ~: ~: ~i~: i ~: ~i i~i .......................................................... ~ ~ iiiii!iiiiiiiiii"":'::::::i:iii~iii~iii~ii!iiE~::~ :::::::::::::::::::::::::::::::: ' '':'::::::i:i:i:? "~

.

~ -

~

......

i ii .~i!i

i::::i!:::::::::::::::::::::::::::::::::::::: ::i::i::::i::iii.~,

::::iii

!~iiiiii!!!!i!i~! i iii!!~iiiii~!iiii~!iii!i!~!~!i!i!~!ii~i~

\

..ill!!!

iii~iii~i~iiii~iii~i~iiiiiii!i!ii!~ii~ii!!!i~i~!ii~i!i!i~i!i~ii!!i!iii~iiiii~!~!~i!i!iii X: interphase- fibrillar

/

i:i:!:i:!:i:i:i:ii:!:i:i:i:i:!:i:iiiii:i:i:iii:iiiiliiii:iiiiiiiiiiii!i!i! iii~iii~!i!i!~i~!i!i!i!i!iiiii!iii!i!i! i!i!i!iii!iii!i!i!!iiill!ii!i

i::::::::i:::: : :::::::::::::::::~::::::~::~i~::~::::ii::i::ii~::~i~i~::~i~~iy :::: i:~i:?:! ~.".-'?.~.:'~ k~-.'~,"-! ~

~::::::::

~.,,:~,~:g:".~ ~.~:~.-.':~::~:,.-~'~:~ i i ~~ "'". ~~.'.-:.-~.x~: E.'-":~::':~:~:~:~:~:~:!*~:~~-..:ii~

li!:.~:~-.-.~.:.~~iil ~ ]~ ~ i .,.~:~..-..

~'."~:~:.

.:::::.~!~N

X and Y: mitosis :::::::::::::::::::::::::::::::: .~ ::~:::~::: ::::::::::::::::::::::::::::::::::::

"~~::~i

iii::!::iii::i::i::iii i::iii::iiiiiii::i::iii::!::i::!::i:?:!::!::!::i~~i!i::iii::il :::::::::: :::::::::::::::::::::::::

""

.-::~:!:::!:i:i:

~iiiiiiii~iiiiiiiiiiii~i~i~i~!~!~i~ii~i~i!iiiiii~i~i~iii!~iiii~iiiiiiiiii~i~ii~iiiiiiiiii~ii~ Y: interphase - non-fibrillar Fig. 3. S c h e m a t i c d r a w i n g of differences in GR-distrihution in m a m m a l i a n cells. Different cell types, x and y indicated in the figure, exhibit a similar GR-pattern during mitosis, but different pattern during interphase, apart f r o m centriolar and granular nuclear staining. GR is indicated w i t h black and dark grey color. The nucleus is represented by a square.

It is important to keep the various staining conditions constant when comparing two methods. It has been reported that when the p H was dropped below 5, even for short periods of time, the staining of the glucocorticoid receptor became nuclear [106]. T h e r e fore, our experiments were all performed at p H 7.35 in PBS-buffer [91].

that the fibroblasts contained specific, saturable and high affinity GR-binding. Whole cell binding assays followed by Scatchard analysis revealed around 100,000 hormone binding sites/cell. Autoradiography after incubating intact monolayer fibroblasts with [3H]dexamethasone me,sylate, which covalently binds to G R [117], showed one band of MR 94,000. Thus, we confirmed many previous investigations showing that human fibroblasts contain one GR-binding species.

only one band is detected, there is a fair probability that at least soluble or extractable contaminating antigens will not be detected by immunocytochemistry [120]. Four different monoclonal anti-GR antibodies, which recognize four different epitopes within the N-terminal regulatory domain of G R [ 104], showed the same Mg94,000-band on Western blots and very similar GR-distributions in cells [47]. This strongly indicates that the staining correctly depicts the actual cellular distribution of GR. Preincubation. Preincubation of the anti-GR antibody " m a b T " with purified rat liver G R blocked the immunostaining efficiently in fixed cells [10, 31, 44, 111]. Preincubation of anti-GR antibodies with molar excess of purified bovine brain M T - p r o t e i n , containing 80% tubulin and 20% MAPs did not appreciably reduce the GR-staining intensity. Preincubation with nonimmune serum from the animal in which the second antibody was raised did not change the staining signal.

Primary antibody specificity

Staining procedure specificity

We had a.ccess to several monoclonal mouse-anti-rat liver G R antibodies, previously produced in our laboratory and shown to recognize both non-activated and activated G R [104, 118]. These antibodies are highly specific for the glucocorticoid receptor and cross-react well with human G R [47, 91,119]. Western immunoblot. A good control when using mono- or poly-clonal antibodies is to perform Western immunoblot experiments on crude cellular extracts. If

Antibody accessibility. It is necessary to test that the method of fixation/permeabilization used for localization studies in a particular cell-system really allows access of the antibodies to all cellular compartments and that it gives reproducible results. We therefore performed control experiments confirming the well established distributions of components in the cytoplasm (various cytoskeletal networks) and the nucelus (nuclear antigens labelled by polyclonal antibodies in

Specificity Ligand binding specificiey Ligand binding analysis in vivo and in vitro showed

Gunnar Akner et al.

10

iiiiiiii i!i ii i!iii!i!i i!i i i i!i i!i i!i iiiiiiiii !i i!i!ji i ii i!i i i i i i!iiiiiiii i

:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

::::.:.::.:.:.::::

:::::::::::::::::::::::

.:,

~!ii~i~i~ii~i!~iii!!~i;!!~iii~iii~iii:i~i~i~i!:~i~i~i:i::i:i~ii~!~i1~!i~!!!ii:~ii:i~

~iiiiiiiiiii: iiiiiiii!iiiiiliiiiDii~!i ii!iiiiiiiiiiiiiiiiiiii iiii!ilii!iiiiiiiliii~i!iiii!ii iiilil iiiiiiiiiiiiliiii!~i!il!iiiiiiiiiii

colchicine

vinblastine

Fig. 4. Schematic drawing of parallel redistribution of GR and tubulin in mammalian cells. Redistribution of GR and tubulin during artificially induced MT-depolymerization. GR is found in cell protrusions and in vinblastine-induced paracrystals and is indicated with black and dark grey color. The nucleus is represented by a square. pooled sera f r o m patients with autoantibodies to nuclear antigens). Both fixation techniques revealed essentially the same results. T h e s e controls showed that the results of immunolocalizing G R would at least not be h a m p e r e d by problems of antibody accessibility. Substitution. Substitution of the first or second or both the first and second antibodies with buffer showed esentially no staining signal in any of the detection systems. T h i s demonstrates that the immunosignal is dependent on the p r i m a r y antibody. Detergent. Including 0.1% T r i t o n X-100 Ce in antibody incubations and washes did not change the staining signal. Limiting dilution. Dilution of the first a n t i - G R antibodies from 300 to 1 0 # g / m l gave rise to the same relative cellular GR-distribution, i.e. no part of the GR-signal could be selectively diluted away.

C L S M provides laser beam excitation of the specimen with a separate monochromatic wavelength for each fluorochrome, compared to a wavelength excitation interval in conventional microscopy. T h i s m a y constitute an advantage in doublestaining experiments. Using C L S M we quantified the immunoreactivity of G R in the cytoplasm and cell nucleus + h o r m o n e treatment after different fixations on thin optical fibroblast sections and analysed the results statistically. T h i s photometric quantification of G R represents the first attempt to actually measure G R immunoreactivity in a larger n u m b e r of cells. T h i s strongly reduces the bias regarding both the selection of cell-fields of view for presentation as well as the description of the visual analysis of GR-distribution. For a detailed description of the C L S M - p h o t o m e t r y process see [47]. A s u m m a r y of our results regarding GR-distribution is presented in T a b l e 5 and in Figs 2-4.

Fluorochrome separation We focused part of the work on assuring reliable fluorochrome separation. Double staining experiments gave the same results as monostainings, both for conventional microscopy and confocal laser scanning microscopy.

Confocal Laser Scanning Microscopy (CLSM) C o m p a r e d to conventional transmission light microscopy, C L S M offers several advantages: (1) C L S M provides thin optical (single or serial) sections of the specimen, avoiding the risk of projection artifacts and also presenting all components in the particular focal plane in focus, regardless of size of the component (e.g. interphase and mitotic cells); (2) C L S M provides better resolution than conventional microscopy, especially along the optical axis, but also laterally; (3) C L S M presents data in a digitalized form which directly allows various kinds of image analysis such as subtraction of one image from another one or quantification, such as m e a s u r e m e n t of fluorescence intensities in various c o m p a r t m e n t s or size of cells or individual compartments; (4) C L S M allows for two- or three-dimensional reconstruction of cells from serial sectioning data; (5)

L O C A L I Z A T I O N OF GR D U R I N G T H E CELL CYCLE

Interphase General. In all m a m m a l i a n cell types examined (i.e. h u m a n p r i m a r y culture fibroblasts from gingiva or skin, h u m a n isolated thymocytes and peripheral blood lymphocytes, mouse spleen lymphocytes and several cell lines, e.g. mouse 3T3, mouse L929, h u m a n H e L a , mouse M C F - 7 and rat H T C ) , G R was distributed in both the cytoplasm and cell nucleus. P h o t o m e t r y on optical sections of h u m a n fibroblasts revealed that 10-12% of the whole cell G R was localized in the nucleus during control culture conditions [47]. T h i s relation was independent of crosslinking or precipitating fixation. Regarding G R , the average GR-staining intensity was ~ 2 . 5 x higher after crosslinking than precipitating fixation, p r e s u m a b l y due to different degrees of fixation-induced extraction of GR. Similar results have been reported when these two types of fixations were compared regarding the immunostaining intensity for nuclear phosphoprotein p105 [121] and SV40 large T - a n t i g e n [122].

Subcellular Distribution of Glucocorticoid Receptor

Nucleus. G R was found to display a diffuse distribution in most interphase nuclei, with an additional granular appearance in a fraction of the nuclei. G R was not present in the nuc1Leoli. Cytoplasm. In the cytoplasm, G R was found to exhibit a fibrillar or non-fibrillar staining pattern depending on cell type. In some cell types, e.g. human fibroblasts of different origin and mouse 3T3-cells, G R exhibited a fibrillar pattern which, in double-staining experiments, colocalized well with tubulin. G R was distributed in a granular pattern along individual fibrils and there was a predilection of G R for a subset of M T s . Subtraction analysis by C L S M further substantiated the close association between G R and tubulin. G R was also found in a more diffuse pattern in the perinuclear area in many cells. In other cell-types, e.g. rat liver HTC-cells and L-cells, the cytoplasmic G R was predominantly diffusely d!istributed in the interphase cell. Plasma membrane. G R also stained parts of the plasma membrane, including vesicle (pinocytotic vacuoles?, lysosomes?) membranes of various sizes, often located along the leading edge of the cells. Cell division G R was located in the mitotic spindle apparatus, both in the pericentriolar area at the spindle poles and along the spindle MTs: kinetochore, astral and possibly also polar M T s . T h e r e was also a diffuse GR-staining outside the mitotic spindle throughout the mitotic cell. Double stainings showed that G R colocalized well with mitotic M T s during all[ stages of mitosis. Possibly, G R distributed in a larger zone than tubulin around the spindle poles. HETEROGENEITY We observed a strong inter- and intra-cellular G R heterogeneity. Such he.terogeneity has previously been described for several :steroid hormone receptors, e.g. G R [43, 106, 123], E R [57, 124] and PR [120]. T h e heterogeneity concern,; both GR-localization and G R intensity and may represent both various cell cycle phases or genetic heterogeneity among the cells. This could probably serve ~LSone explanation for the sometimes reported lack of good correlation between glucocorticoid dose, GR-quantity and cellular response. DISTRIBUTION OF GR A N D hsp90 A F T E R DEPOLYMERIZATION OF M T S

Drug-induced depolymerization 10/~M colchicine induced an almost complete depolymerization of M T s within 1 h. Both g- and fltubulin were distributed diffusely over the whole cell, leaving only occasional M T s intact. Both tubulin isoforms also localized in newly formed plasma membrane processes, known to contain cell organelles such as lysosomes, ribosomes and mitochondria [125]. Simul-

11

taneously, G R was reorganized in a very similar manner, indicating that G R is associated with individual tubulin dimers. T h e nuclear GR-staining remained unchanged after MT-depolymerization and G R was not observed in occasional intact M T s . Depolymerization using 10 # M vinblastine or 10 p M nocodazole showed essentially the same effect as colchicine, but vinblastine induced GR-containing paracrystals as well [91, 92]. T r e a t m e n t with 1 0 # M cytochalasin B for 1-2 h induced a strong arborization of the cells. Staining for G R showed a strong resemblance to the tubulin pattern, even though the individual M T s were not easily discernible after the strong morphological derangement. After cytochalasin B treatment, the actin staining pattern differed from that of G R and tubulin.

Cold-induced depolymerization Exposure of cultured fibroblasts to + 4 ° C for 2.5 h induced an almost complete depolymerization of cytoplasmic M T s with a parallel change from fibrillar to mainly diffuse GR-staining. However, the typical features of d r u g q n d u c e d MT-depolymerization and formation of new plasma membrane processes containing G R and tubulin, were not observed after cold-induced depolymerization. One possible explanation for this phenomenon may be that cold treatment reduces, whereas MT-inhibitors increase cytoplasmic mass flow [1251. T R E A T M E N T WITH GLUCOCORTICOID HORMONES

We have never observed any distinct compartment shift of G R from cytoplasm to the cell nucleus after glucocorticoid hormone treatment in any of the mammalian cells tested, regardless of the type of fixation/permeabilization, cell culture conditions, or glucocorticoid administration. T h e rather large G R heterogeneity, however, made it difficult to visually determine a possible, small hormone-induced change in GR-distribution. We therefore quantified the photometric GR-intensities in the nucleus and cytoplasm on thin, C L S M produced optical sections of human fibroblasts monostained for GR, using two standard fixations, with or without treatment with glucocorticoid [47]. This analysis revealed a hormone-induced significant increase in GR-immunoreactivity in both the nucleus and cytoplasm compared to controls. Since these effects were only detected after precipitating but not after crosslinking fixation, the results were interpreted as evidence in support of a hormone-induced increase in GR-affinity to existing docking sites in both nucleus and cytoplasm, without any sign of intracellular compartment shift. This change in GR-affinity may give rise to a visual impression of a partial nuclear translocation in some cells.

Gunnar Akner et al.

12

Similar results were obtained regardless of whether the GR-intensities in the nucleus and cytoplasm or the quotient between them were expressed per whole cell or per pixel, thereby excluding the possible influence by hormone induced change in the size of the whole cell or of the individual compartment(s). Similar results of GR-localization after hormone treatment were obtained after preceding drug-induced disassembly of M T s as well as after inhibition of energy synthesis by drugs such as oligomycin or Na-azide. However, these observations were not analysed by photometry. TREATMENT WITH HEAT STRESS

Heat shock did not appreciably affect the cellular distribution of G R on visual analysis, but induced a reversible nuclear translocation of hsp90 [126]. T h i s finding contrasts with a recent report claiming that heat shock induces a nuclear translocation of G R in vitro, i.e. G R is detected in the cytosol before and in the pellet fraction after heat shock treatment [127]. T h e discrepancy may be explained by a similar reasoning as applied to the glucocorticoid induced nuclear translocation in vitro described above. A T T E M P T S TO ASSESS A F U N C T I O N A L S I G N I F I C A N C E OF A G R - M T I N T E R A C T I O N During the course of this study, we have tried a n u m b e r of in vivo assays to analyse the putative functional significance of the observed interaction between G R and M T s . We hypothesized that glucocorticoidregulated function(s) may depend on an intact interaction between G R and M T s . I f this was the case, glucocorticoids would not be able to elicit their effects after M T - d e p o l y m e r i z a t i o n . T h i s hypothesis has been tested for the following glucocorticoid test-systems: (i) induction of tyrosine aminotransferase ( T A T ) enzyme activity in rat liver H T C - c e l l s ; (ii) induction of alkaline phosphatase (ALP) enzyme activity in h u m a n fibroblasts; (iii) inhibition of the release of [3H]arachidonic acid from h u m a n fibroblasts preincubated with [3H]arachidonic acid; and (iv) inhibition of uptake of [aH]thymidine in h u m a n fibroblasts. For several tested assays, we observed that the M T - d r u g s in the tested doses (1-10/~M) by themselves affected the tested variable to a large degree, sometimes more than the glucocorticoids alone. N o n e of these in vivo assays revealed any consistent difference between glucocorticold-induced effects with intact or depolymerized MTs. T h e r e are several possible reasons for this. (a) T h e tested parameters rely on GR-action at the nuclear genome and such effects may not depend on intact M T s . It is possible that other glucocorticoid functions, that do not involve nuclear genomic regulation, are M T - d e p e n d e n t . (b) Depolymerization of M T s using 10 ~ M colchicine or vinblastine leaves a small n u m b e r

of assembled M T s that are resistant to drug-treatment. Even though we have not observed any GR-staining in these drug-resistant M T s , this residual M T - p o p u lation may be sufficient for transduction of the glucocorticoid/GR-effect(s). (c) G R and M T s are not colocalized in all cell types. (d) T h e interaction between G R and M T s is coincidental or non-functional. S P E C U L A T I O N S ON F U N C T I O N A L S I G N I F I C A N C E OF A G R - M T I N T E R A C T I O N Based on some of the observations described in this review, there are several hypothetical possibilities as to how the effect of glucocorticoids may be transduced to cells, involving the well described effects of G R in the nucleus alone as well as extranuclear GR.

Extranuclear G R Glucocorticoid effects m a y be transduced through the G R indirectly or directly on site in the cytoplasm, without involving the nuclear genome.

Indirect effects Glucocorticoids/GR m a y operate at the mitochondrial genome or at the putative centrosomal genome.

Direct effects M T s . Glucocorticoids/GR may exert direct, extragenomic effects on M T s : (a) glucocorticoids m a y regulate cytoplasmic and/or mitotic M T s directly by G R being a M A P or by G R binding to a M A P ; (b) M T s m a y regulate cytoplasmic and/or mitotic G R in some as yet unidentified way(s). Centrosomes. Several steroid hormones are reported to be able to bind to the centrosome [57] possibly because the corresponding receptors are present in this organelle, in a similar way as G R described here. T h i s may constitute a mechanism for steroids to directly affect the centriolar cycle and thereby e.g. cell growth. Lysosomal membranes. Glucocorticoids have been reported to stabilize lysosomal m e m b r a n e s by a mechanistically unknown process [128]. It is possible that this effect is mediated directly in the lysosomal m e m b r a n e via m e m b r a n e - b o u n d GR. T h e r e is evidence that after glucocorticoid-induced treatment in vitro, only 3 5 - 6 0 % of the dissociated G R is activated and thus exhibits D N A - b i n d i n g capacity [129]. T h e remaining 4 0 - 6 5 % n o n - D N A binding G R pool has a more acid p I than the D N A - b i n d i n g G R pool. T h e authors proposed that these two G R - p o o l s represent different energy states of folding after dissociation of hsp90. T h e function of the n o n - D N A binding but dissociated G R is unknown, but may hypothetically be related to some glucocorticoid effect(s) in the cytoplasm. Nuclear G R Besides nuclear G R participating in transcriptional regulation of specific target genes, granular nuclear G R

Subcellular Distribution of Glucocorticoid Receptor m a y be l o cal i zed in small r i b o n u c l e o p r o t e i n ( s n R N P ) particles. T h e r e is e v i d e n c e f r o m a c o n f o c a l laser m i c r o s c o p i c analysis t h a t o v e r e x p r e s s e d h e t e r o l o g o u s G R is l o cal i zed in a n o n - r a n d o m m a n n e r in n u c l e i in a p a t t e r n r e s e m b l i n g ~:hat o f s n R N P s [130]. O t h e r s t e r o i d h o r m o n e - r e c e p t o r c o m p l e x e s , i.e. E R an d A R , h a v e b e e n r e p o r t e d to be associated w i t h s n R N P s [131]. Such snRNP-particles exhibit a granular distribution p a t t e r n in i n t e r p h a s e n u c l e i [132, 133]. S e v e r a l different s n R N P s p a r t i c i p a t e in th e p r o c e s s i n g o f n e w l y f o r m e d m R N A [134]. T a k e n t o g e t h e r , this m i g h t i m p l y that G R is associated w i t h s n R N P s a n d t h e r e b y p a r t i c i pates in p o s t - t r a n s c r i p t i o n a l m R N A m a t u r a t i o n . GR In The Mitotic Spindle

M i t o t i c s p i n d l e G R m a y be i n v o l v e d in t r a n s d u c i n g g l u c o c o r t i c o i d effect(s) d i r e c t l y to s p i n d l e M T s t h e r e b y e x e r t i n g its well d o c u m e n t e d g r o w t h - m o d u l a t i n g effects. CONCLUDING REMARKS T h e r e is e v i d e n c e t h a t G R is associated w i t h t h e cytoskeleton, both the microfilament- and M T - n e t works. R e g a r d i n g M T s , G R is c o l o c a l i z e d w i t h t u b u l i n d u r i n g m i t o s i s in all a n d d u r i n g i n t e r p h a s e in s o m e o f t h e i n v e s t i g a t e d m a m m a l i a n cell types. Besides its well k n o w n n u c l e a r d i s t r i b u t i o n , G R s e e m s to be u n i q u e a m o n g t h e v a r i o u s p r o t e i n s in th e s t e r o i d h o r m o n e r e c e p t o r s u p e r f a m i l y in also h a v i n g a d i s t i n c t c y t o p l a s m i c l o cat i o n , w h e r e it associates w i t h several parts o f th e c y t o s k e l e t o n as well as w i t h d i f f e r e n t c y t o p l a s m i c organelles. E v i d e n c e f o r an i n t e r a c t i o n w i t h M T s has also b e e n p r e s e n t e d f o r a n o t h e r m e m b e r o f t h e st er o i d r e c e p t o r s u p e r f a m i l y , i.e. th e 1 , 2 5 - d i h y d r o x y - v i t a m i n D r e c e p t o r [115]. S t u d i e s are in p r o g r e s s in o u r l a b o r a t o r y to f u r t h e r analyse w h e t h e r the s t r u c t u r a l G R - M T i n t e r a c t i o n is p h y s i o l o g i c a l l y r e l e v a n t for cells an d tissues. Acknowledgement--This work: was supported by a grant from the Swedish Medical Research Council, No. 13X-2819.

REFERENCES 1. Wrange 0. and Gustafsson J.-A.: Separation of the hormoneand DNA-binding sites of the hepatic glucocorticoid receptor by means of proteolysis. J. Biol. Chem. 253 (1978) 856-865. 2. Wrange 0., Okret S., Radojcic M., Carlstedt-Duke J. and Gustafsson J.-A.: Characterization of the purified activated glucocorticoid receptor from rat liver cytosol. J. Biol. Chem. 259 (1984) 4534-4541. 3. Evans R. M.: The steroid and thyroid hormone receptor superfamily. Science 240 (1988) 889-895. 4. Miller J., McLachlan A. D. and Klug A.: Repetitive zinc-binding domains in the protein transcription factor IIIA from Xenopus oocytes. E M B O J. 4 (1985) 1609-1614. 5. Evans R. M. and Hollenberg S. M.: Zinc fingers: gilt by association. Cell 52 (1988) 1-3. 6. GigOere V., Hollenberg S. M., Rosenfeld M. G. and Evans R. M.: Functional domains of the human glucocorticoid receptor. Cell 46 (1986) 645-652.

13

7. Housley P. R., Sanchez E. R., Danielsen M., Ringold E. M. and Pratt W. B.: Evidence that the conserved region in the steroid binding domain of the glucocorticoid receptor is required for both optimal binding of hsp90 and protection from proteolytic cleavage. J. Biol. Chem. 265 (1990) 12,778-12,781. 8. Godowski P. J., Rusconi S., Miesfeld S. and Yamamoto K. R.: Glucocorticoid receptor mutants are constitutive activators of transcriptional enhancement. Nature 325 (1987) 365-368. 9. Rusconi S. and Yamamoto K. R.: Functional dissection of the hormone and the DNA binding activities of the glucocorticoid receptor. E M B O J. 6 (1987) 1309-1315. 10. Picard D. and Yamamoto K. R.: Two signals mediate hormonedependent nuclear localization of the glucocorticoid receptor. E M B O J. 6 (1987) 3333-3340. 11. Markland W., Smith A. E. and Roberts B. L.: Signal-dependent translocation of simian virus 40 large-T-antigen into rat liver nuclei in a cell-free system. Molec. Cell. Biol. 7 (1987) 4255-4265. 12. Hollenberg S. M., Weinberger C., Ong E. S., Cerelli (3., Oro A., Lebo R., Thompson E. B., Rosenfeld M. G. and Evans R. M.: Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 318 (1985) 635-641. 13. Carlstedt-Duke J., Str6mstedt P. E., Wrange 0., Bergman T., Gustafsson J.-A. and J6rnvall H.: Domain structure of the glucocorticoid receptor protein. Proc. Natn. Acad. Sci. U.S.A. 84 (1987) 4437-4440. 14. Denis M., Gustafsson J.-A. and Wikstr6m A.-C.: Interaction of the Mr90,000 heat shock protein with the steroid-binding domain of the glucocorticoid receptor. J. Biol. Chem. 263 (1988) 18,520-18,523. 15. Sanchez E. R., Faber L. E., Henzel W. J. and Pratt W. B.: The 56-59-kilodalton protein identified in untransformed steroid receptor complexes is a unique protein that exists in cytosol in a complex with both the 70- and 90-kilodalton heat shock proteins. Biochemistry 29 (1990) 5145-5152. 16. Tai P. K. K., Albers M. W., Chang H., Faber L. E. and Schreiber S. L.: Association of a 59-kilodalton immunophilin with the glucocorticoid receptor complex. Science 256 (1992) 1315-1318. 17. Renoir J. M., Radanyi C., Faber L. E. and Baulieu E. E.: The non-DNA-binding heterooligomeric form of mammalian steroid hormone receptors contains a hsp90-bound 59-kilodalton protein. J. Biol. Chem. 265 (1990) 10,740-10,745. 18. Economidis I. V. and Rousseau G. G.: Association of the glucocorticoid hormone receptor with ribonucleic acid. F E B S Lett. 181 (1985) 47-52. 19. Miller-Diener A., Schmidt T. J. and Litwack G.: Protein kinase activity associated with the purified rat hepatic glucocorticoid receptor. Proc. Natn. Acad. Sci. U.S.A. 82 (1985) 4003-4007. 20. Munck A.: Glucocorticoid inhibition of glucose uptake by peripheral tissues: old and new evidence, molecular mechanisms, and physiological significance. Perspect. Biol. Med. 14 (1971) 265-289. 21. Wrange 0., Carlstedt-Duke J. and Gustafsson J.-A.: Purification of the glucocorticoid receptor from rat liver cytosol. ft. Biol. Chem. 254 (1979) 9284-9290. 22. Eriksson P., Daneholt B. and Wrange 0.: The glucocorticoid receptor in homodimeric and monomeric form visualized by electron microscopy. J. Struct. Biol. 107 (1991) 48-55. 23. Okret S., Carlstedt-Duke J., Wrange (3., Carlstr6m K. and Gustafsson J.-A.: Characterization of an antiserum against the glucocorticoid receptor. Biochim. Biophys. Acta 677 (1981) 205-219. 24. Wrange 0., Eriksson P. and Perlmann T.: The purified activated glucocorticoid receptor is a homodimer. J. Biol. Chem. 264 (1989) 5253-5259. 25. Antakly T., Sasaki A., Liotta A. S., Palkovits M. and Krieger D. T.: Induced expression of the glucocorticoid receptor in the rat intermediate pituitary lobe. Science 229 (1985) 277-279. 26. Antakly T. and Eisen H. J.: Immunocytochemical localization of glucocorticoid receptor in target cells. Endocrinology 15 (1984) 1984-1989. 27. Ballard P. L., Baxter J. D., Higgins S. J., Rousseau G. G. and Tomkins G. M.: General presence of glucocorticoid receptors in mammalian tissues. Endocrinology 94 (1974) 998-1002. 28. Farman N., Oblin M. E., Lombes M., Delahaye F., Westphal H. M., Bonvalet J. P. and Gasc J. M.: Immunolocalization of

14

29.

30.

31.

32. 33.

34.

35. 36.

37. 38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48. 49.

G u n n a r A k n e r et al. gluco-mineralocorticoid receptors in rabbit kidney. Am. J. Physiol. (Cell Physiol. 29) 260 (1991) C226-C233. Antakly T., Zhang C., Sarrieau A. and Raquidan D.: Cellspecific expression of the glucocorticoid receptor within granular convoluted tubules of the rat submaxillary gland. Endocrinology 128 (1991) 617-622. Teasdale J., Lewis F. A., Barrett I. D., Abbott A. C., Wharton J. and Bird C. C.: Immunocytochemical application of monoclonal antibodies to rat liver glucocorticoid receptor. J. Pathol. 150 (1986) 227-237. Fuxe K., Wikstr6m A.-C., Okret S., Agnati L. F., H~irfstrand A., Yu Z.-Y., Granholm L., Zoli M., Vale W. and Gustafsson J.-A.: Mapping of glucocorticoid receptor immunoreactive neurons in the rat tel- and diencephalon using a monoclonal antibody against rat liver glucocorticoid receptor. Endocrinology 117 (1985) 1803-1812. Gametchu B.: Glucocorticoid receptor-like antigen in lymphoma cell membranes: correlation to cell lysis. Science 236 (1987) 456-461. Jensen E. V., Suzuki T., Kawashima T., Stumpf W. E., Jungblut P. W. and DeSombre E. R.: A two-step mechanism for the interaction of estradiol with rat uterus. Proc. Natn. Acad. Sci. U.S.A. 59 (1968) 632-638. Grody W. W., Schrader W. T. and O'Malley B. W.: Activation, transformation and subunit structure of steroid hormone receptors. Endocr. Rev. 3 (1982) 141. Gorski J. and Gannon F.: Current models of steroid hormone action: a critique. Ann. Rev. Physiol. 38 (1976) 425-450. Sheridan P. J., Buchanan J. M., Anselmo V. C. and Martin P. M.: Equilibrium: the intracellular distribution of steroid receptors. Nature 282 (1979) 579-582. Martin P. M. and Sheridan P. J.: Intracellular distribution of estrogen receptors: a function of preparation. Experientia 36 (1980) 620-622. Martin P. M. and Sheridan P. J.: Intracellular localization of steroid hormone receptors. A challenging problem. Autoradiographic and biochemical study. Ann. Pathol. 6 (1986) 115-129. Guiochon-Mantel A., Lescop P., Christin-Maitre S., Loosfelt H., Perrot-Applanat M. and Milgrom E.: Nucleocytoplasmatic shuttling of the progesterone receptor. E M B O J. 10 (1991) 3851-3859. Perrot-Applanat M., Lescop P. and Milgrom E.: The cytoskeleton and the cellular traffic of the progesterone receptor, ft. Cell. Biol. 119 (1992) 337-348. King W. J. and Greene G. L.: Monoclonal antibodies localize estrogen receptor in the nuclei of target cells. Nature 307 (1984) 745-747. Sanchez E. R., Hirst M., Scherrer L. C., Tang H. Y., Welsh M. J., Harmon J. M., Simons Jr S. S., Ringold G. M. and Pratt W. B.: Hormone-free mouse glucocorticoid receptors overexpressed in chinese hamster ovary cells are localized to the nucleus and are associated with both hsp70 and hsp90. J. Biol. Chem. 265 (1990) 20,123-20,130. LaFond R. E., Kennedy S. W., Harrison R. W. and Villee C. A.: Immunocytochemical localization of glucocorticoid receptors in cells, cytoplasts and nucleoplasts. Exp. Cell Res. 175 (1988) 52-62. Gasc J.-M., Delahaye F. and Baulieu E. E.: Compared intracellular localization of the glucocorticosteroid and progesterone receptors: an immunocytochemical study. Exp. Cell Res. 181 (1989) 492-504. Brink M., Humbel B. M., De Kloet E. R. and Van Driel R.: The unliganded glucocorticoid receptor is localized in the nucleus, not in the cytoplasm. Endocrinology 130 (1992) 3575-3581. Pekki A., Koistinaho J., Ylikomi T., Vilja P., Westphal H. and Touhimaa P.: Subcellular location of unoccupied and occupied glucocorticoid receptor by a new immunohistochemical technique. J. Steroid Biochem. Molec. Biol. 41 (1992) 753-756. Akner G., Wikstr6m A.-C., Mossberg K., Sundqvist K.-G. and Gustafsson J.-A.: Morphometric studies of the localization of the glucocorticoid receptor in mammalian cells and of glucocorticoid hormone-induced effects. J. Histochem. Cytochem. 42 (1994) 645-657. Paine P. L., Moore L. C. and Horowitz S. B.: Nuclear envelope permeability. Nature 254 (1975) 109. Feldherr C. M., Kallenbach E. and Schultz N.: Movement of

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62. 63.

64.

65.

66. 67. 68.

69. 70. 71.

72.

a karyophilic protein through the nuclear pores of oocytes. ft. Cell Biol. 99 (1984) 2216-2222. LaCasse E. C. and Lefebvre Y. A.: Nuclear and nuclear envelope binding proteins of the glucocorticoid receptor nuclear localization peptide identified by crosslinking. J. Steroid Biochem. Molec. Biol. 40 (1991) 279-285. Carson-Jurica M. A., Schrader W. T. and O'Malley B. W.: Steroid receptor family: structure and functions. Endocr. Rev. 11 (1990) 201-220. Dong Y., Poellinger L., Gustafsson J.-A. and Okret S.: Regulation of glucocorticoid receptor expression: evidence for transcriptional and posttranscriptional mechanisms. Molec. Endocrinol. 2 (1988) 1256-1264. McIntyre W. R. and Samuels H. H.: Triamcinolone acetonide regulates glucocorticoid-receptor levels by decreasing the halflife of the activated nuclear-receptor form. J. Biol. Chem. 260 (1985) 418-427. Munck A., Mendel D. B., Smith L. I. and Orti E.: Glucocorticoid receptors and actions. Am. Rev. Respir. Dis. 141 (1990) $2-S10. Ichii S. and Yoshida A.: Effect of colchicine on the depletion and replenishment of cytoplasmic glucocorticoid receptor in rat liver after administration of glucocorticoid. Endocrinol. ffapon 32 (1985) 225-231. Alberts B., Bray D., Lewis J., Raft M., Roberts K. and Watson J. D. (eds): Molecular Biology of the Cell. Garland Publ. Inc. (1989). Nenci I. and Marchetti E.: Concerning the localization of steroids in centrioles and basal bodies by immunofluorescence. J. Cell Biol. 76 (1978) 255-260. Miyata Y. and Yahara I.: Cytoplasmic 8S glucocorticoid receptor binds to actin filaments through the 90-kDa heat shock protein moiety, ft. Biol. Chem. 266 (1991) 8779-8783. Castellino F., Heuser J., Marchetti S., Bruno B. and Luini A.: Glucocorticoid stabilization of actin filaments: a possible mechanism for inhibition of corticotropin release. Proc. Natn. Acad. Sci. U.S.A. 89 (1992) 3775-3779. Clark A. F., Wilson K., McCartney M. D., Miggans S. T., Kunkle M. and Howe W.: Glucocorticoid-induced formation of cross-linked actin networks in cultured human trabecular meshwork cells. Invest. Ophtalmol. 35 (1994) 281-294. Wilson L.: Microtubules as drug receptors: pharmacological properties of microtubule protein. Ann. N. Y. Acad. Sci. 253 (1975) 213-231. Dustin P.: Microtubules. Springer-Verlag, Berlin (1984). Hall J. L., Dudley L., Dobner P. R., Lewis S. A. and Cowan N. J.: Identification of two human fl-tubulin isotypes. Molec. Cell Biol. 3 (1983) 854-862. Cowan N. J., Dobner P. R., Fuchs E. V. and Cleveland D. W.: Expression of human ~t-tubulin genes: interspecies conservation of 3' untranslated regions. Molec. Cell Biol. 3 (1983) 1738-1745. Hiller G. and Weber K.: Radioimmunoassay for tubulin: a quantitative comparison of the tubulin content of different established tissue culture cells and tissues. Cell 14 (1978) 795-804. Wiche G., Oberkanin C. and Himmler A.: Molecular structure and function of microtubule-associated proteins. Int. Rev. Cytol. 124 (1991) 217-273. Brinkley B. R.: Microtubule organizing centers. Ann. Rev. Cell Biol. 1 (1985) 145-172. Bennett V. and Davis J.: Erythrocyte ankyrin: immunoreactive analogues are associated with mitotic structures in cultured cells and with microtubules in brain. Proc. Natn. Acad. Sci. U.S.A. 78 (1981) 7550-7554. Cleveland D. W.: Microtubule MAPping. Cell 60 (1990) 701-702. Padilla R., Maccioni R. B. and Avila J.: Calmodulin binds to a tubulin binding site of the microtubule-associated protein tau. Molec. Cell. Biochem. 97 (1990) 35-41. Karasev A. V., Kashina A. S., Gelfand V. I. and Dolja V. V.: Hsp70-related 65 kDa protein of beet yellows closterovirus is a microtubule-binding protein. F E B S Lett. 304 (1992) 12-14. Duerr A., Pallas D. and Solomon F.: Molecular analysis of cytoplasmic microtubules in situ: identification of both widespread and specific proteins. Cell 24 (1981) 203-211.

Subcellular D i s t r i b u t i o n o f G l u c o c o r t i c o i d R e c e p t o r 73. Rodionov V. I., Gyoeva F. K., Kashina A. S., Kuznetsov S. A. and Gelfand V. I.: Microtubule-associated proteins and microtubule-based translocators have different binding sites on tubulin molecule, ft. Biol. Chem. 265 (1990) 5702-5707. 74. Lewis S. A., Wang D. and Cowan N. J.: Microtubuleassociated protein MA2 shares a microtubule binding motif with tau protein. Science 242 (1988) 936-939. 75. De la Torre J., Carrascosa J. L. and Avila J.: The localization of tau proteins on the microtubule surface. Eur..7. Cell Biol. 40 (1986) 233-237. 76. Drubin D. G., Feinstein S. C., Shooter E. M. and Kirschner M. W.: Nerve growth factor-induced neurite outgrowth in PC12 cells involves the coordinate induction of microtubule assembly and assembly..promoting factors..7. Cell Biol. 101 (1985) 1799-1807. 77. Wiche G., Corces V. G and Avila J.: Preferential binding of hog brain microtubule-associated proteins to mouse satellite versus bulk DNA preparations. Nature 273 (1978) 403-405. 78. Sato C., Nishizawa K., Nakamura H., Komagoe Y., Shimada K., Ueda R. and Suzuki S.: Monoclonal antibody against microtubule associated protein-1 produces immunofluorescent spots in the nucleus and centrosome of cultured mammalian cells. Cell Struct. Functwn 8 (1983) 245-254. 79. Connolly J. A., Kalnins V. I., Cleveland D. W. and Kirschner M. W.: Immunofluorescent staining of cytoplasmic and spindle microtubles in mouse fibroblasts with antibody to T protein. Proc. Natn. Acad. Sci. U.S.A. 74 (1977) 2437-2440. 80. Neighbors B. W., Williams Jr R. C. and Mclntosh J. R.: Localization of kinesin irk cultured cells. J. Cell Biol. 106 (1988) 1193-1204. 81. Pfarr C. M., Coue M., Grissom P. M., Hays T. S., Porter M. E. and Mclntosh J. R.: Cytoplasmic dynein is localized to kinetochores during mitosis. Nature 345 (1990) 263-265. 82. Redmond T., Sanchez E. R., Bresnick E. H., Schlesinger M. J., Toft D. O., Pratt W. B. and Welsh M. J.: Immunofluorescence colocalization of the 90-kDa heat-shock protein and microtubules in interphase and mitotic mammalian cells. Eur. J. Cell Biol. 50 (1989) 66-75. 83. Weller N. K.: A 70kDa microtubule-associated protein in NIL8 cells comigrates with the 70 kDa heat shock protein. Biol. Cell 63 (1988) 307-317. 84. Welsh M. J., Dedman J. R., Brinkley B. R. and Means A. R.: Calcium-dependent regulator protein: Localization in mitotic apparatus of eukaryotic cells. Proc. Natn. Acad. Sci. U.S.A. 75 (1978) 1867-1871. 85. Serrano L., Hermindez M. A., Diaz-Nido J. and Avila J.: Association of casein kin~LseII with microtubules. Exp. Cell Res. 181 (1989) 263-272. 86. Browne C. L., Lockwood A. H., Su J. L., Beavo J. A. and Steiner A. L.: Immunofluorescent localization of cyclic nucleotide-dependent protein kinases in the mitotic apparatus of cultured cells. `7. Cell Biol. 87 (1980) 336-345. 87. Guerriero Jr V., Rowley D. R. and Means A. R.: Production and characterization of an antibody to myosin light chain kinase and intracellular localization of the enzyme. Cell 27 (1981) 449-458. 88. Fujiwara K. and Pollard T. D.: Fluorescent antibody localization of myosin in the cytoplasm, cleavage furrow, and mitotic spindle of human cells..t. Cell Biol. 71 (1976) 848-875. 89. Cande W. Z., Lazarides E. and Mclntosh J. R.: A comparison of the distribution of actin and tubulin in the mammalian mitotic spindle as seen b,.g indirect immunofluorescence. J. Cell Biol. 72 (1977) 552-567. 90. Amsterdam A., Pitzer F., Baumeister W.: Changes in intracellular localization of proteasomes in immortalized ovarian granulosa cells during mitosis associated with a role in cell cycle control. Proc. Natn. Acad. Sci. U.S.A. 90 (1993) 99-103. 91. Akner G., Sundqvist K.-G., Denis M., Wikstr6m A.-C. and Gustafsson J.-A.: Immunocytochemical localization of glucocorticoid receptor in human gingival fibroblasts and evidence for a colocalization of glucocorticoid receptor with cytoplasmic microtubules. Eur. ft. Call Biol. 53 (1990) 390-401. 92. Akner G., Mossberg K., '~,rikstr6m A.-C., Sundqvist K.-G. and Gustafsson J.-A.: EviderLce for colocalization of glucocorticoid receptor with cytoplasmic microtubules in human gingival fibroblasts using two different monoclonal anti-GR antibodies,

93.

94. 95. 96. 97.

98.

99.

100.

101.

102.

103.

104.

105. 106.

107.

108.

109. 110. 111.

112.

113.

15

confocal laser scanning microscopy and image analysis. ft. Steroid Biochem. Molec. Biol. 39 (1991) 419-432. Osborn M. and Weber K.: Cytoplasmic microtubules in tissue culture cells appear to grow from an organizing structure towards the plasma membrane. Proc. Natn. Acad. Sci. U.S.A. 73 (1976) 867-871. Wolfe J.: Basal body fine structure and chemistry. Adv. Cell. Molec. Biol. 2 (1972) 151-192. Bornens M., Paintrand M., Berges J., Marty M. C. and Karsenti E.: Structural and chemical characterization of isolated centrosomes. Cell Motil. Cytoskel. 8 (1987) 238-249. Verde F., Labbe J., Doree M. and Karsenti E.: Regulation of microtubule dynamics by cdc2 protein kinase in cell-free extracts of Xenopus eggs. Nature 343 (1990) 233-237. Hall J. L., Ramanis Z. and Luck D. J. L.: Basal body/centriolar DNA: Molecular genetic studies in chlamydomonas. Cell 59 (1989) 121-132. Hirsch P. C. and Szego C. M.: Estradiol receptor functions of soluble proteins from target-specific lysosomes. J. Steroid Biochem. 5 (1974) 533-542. Welshons W. V., Krummel B. M. and Gorski J.: Nuclear localization of unoccupied receptors for glucocorticoids, estrogens and progesterone in GH3 cells. Endocrinology 117 (1985) 2140-2147. Coutard M., Osborne-Pellegrin M. J. and Funder J. W.: Tissue distribution and specific binding of tritiated dexamethasone in vivo: autoradiographic and cell fractionation studies in the mouse. Endocrinology 103 (1978) 1144-1152. Stalker A., Hermo L. and Antakly T.: Covalent affinity labeling, radioautography, and immunocytochemistry localize the glucocorticoid receptor in rat testicular leydig cells. Am..7. Anat. 186 (1989) 369-377. Stumpf W. E. and Sar M.: Steroid hormone sites of action in the brain. In: Steroid Hormone Regulation of the Brain. Proceedings of an International Symposium held at The Wenner-Gren Center, Stockholm, 27-28 October 1980 (Edited by K. Fuxe., J.-A. Gustafsson and L. Wetterberg). Pergamon Press, New York (1981) pp. 41-50. Westphal H. M., Moldenhauer G. and Beato M.: Monoclonal antibodies to rat liver glucocorticoid receptor. E M B O .7. 1 (1982) 1467-1471. Okret S., Wikstr6m A.-C., Wrange 0., Andersson B. and Gustafsson J.-A.: Monoclonal antibodies against the rat liver glucocorticoid receptor. Proc. Natn. Acad. Sci. U.S.A. 81 (1984) 1609-1613. Gametchu B. and Harrison R. W.: Characterization o f a monoclonal antibody to the rat liver glucocorticoid receptor. Endocrinology 114 (1984) 274-279. Antakly T., O'Donnell D. and Thompson E. B.: Immunocytochemical localization of the glucocorticoid receptor in steroidsensitive and -resistant human leukemic cells. Cancer Res. 50 (1990) 1337-1345. Middlebrook J. L., Wong M. D., Ishii D. N. and Aronow L.: Subcellular distribution of glucocorticoid receptors in mouse fibroblasts. Biochemistry 14 (1975) 180-186. Cidlowski J. A. and Munck A.: Multiple forms of nuclear binding of glucocorticoid-receptor complexes in rat thymocytes. J. Steroid Biochem. 13 (1980) 105-112. Ueda K., Isohashi F., Okamoto K., Yoshikawa K. and Sakamoto Y.: Interaction of rat liver glucocorticoid receptor with histones. Endocrinology 124 (1989) 1042-1049. Mithieux G., Alquier C., Roux B. and Rousset B.:'Interaction of tubulin with chromatin proteins. `7. Biol. Chem. 259 (1984) 15,523-15,531. Wikstr6m A.-C., Bakke O., Okret S., Br6nneg~rd M. and Gustafsson J.-A.: Intracellular localization of the glucocorticoid receptor: evidence for cytoplasmic and nuclear localization. Endocrinology 120 (1987) 1232-1242. Pratt W. B., Sanchez E. R., Bresnick E. H., Meshinchi S., Scherrer L. C., Dalman F. C. and Welsh M. J.: Interaction of the glucocorticoid receptor with the Mr90,000 heat shock protein: an evolving model of ligand-mediated transformation and translocation. Cancer Res. 49 (1989) 2222s-2229s. Scherrer L. C. and Pratt W. B.: Association of the transformed glucocorticoid receptor with a cytoskeletal protein complex. `7. Steroid Biochem. Molec. Biol. 41 (1992) 719-721.

16

G u n n a r A k n e r et al.

114. Scherrer L. C. and Pratt W. B.: Energy-dependent conversion of transformed cytosolic glucocorticoid receptors from soluble to particulate-bound form. Biochemistry 31 (1992) 10,879-10,886. 115. Barsony J. and McKoy W.: Molybdate increases intracellular 3', 5'-guanosine cyclic monophosphate and stabilizes vitamin D receptor association with tubulin-containing filaments. J. Biol. Chem. 267 (1992) 24,457-24,465. 116. Baschong W., Baschong-Prescianotto C. and Kellenberger E.: Reversible fixation for the study of morphology and macromolecular composition of fragile biological structures. Eur. J. Cell Biol. 32 (1983) 1-6. 117. Stoney Simons Jr S. and Thompson B.: Dexamethasone 21mesylate: an affinity label of glucocorticoid receptors from rat hepatoma tissue culture cell. Proc. Natn. Aead. Sci. U.S.A. 78 (1981) 3541-3545. 118. Denis M., Wikstr6m A.-C. and Gustafsson J.-A.: The molybdate-stabilized non-activated glucocorticoid receptor contains a dimer of Mr90,000 non-hormone-binding protein. J. Biol. Chem. 262 (1987) 11,803-11,806. 119. Br6nneg~rd M., Poellinger L., Okret S., Wikstr6m A.-C., Bakke O. and Gustafsson J.-A.: Characterization and sequence specific binding to mouse mammary tumor virus DNA of purified activated human glucocorticoid receptor. Biochemistry 26 (1987) 1697-1704. 120. Perrot-Applanat M., Logeat F., Groyer-Picard M. T. and Milgrom E.: Immunocytochemical study of mammalian progesterone receptor using monoclonal antibodies. Endocrinology 116 (1985) 1473-1484. 121. Clevenger C. V., Bauer K. and Epstein A. L.: A method for simultaneous nuclear immunofluorescence and DNA content quantitation using monoclonal antibodies and flow cytometry. Cytometry 6 (1985) 208-214. 122. Schimenti K. J. and Jacobberger J. W.: Fixation of mammalian cells for flow cytometric evaluation of DNA content and nuclear immunofluorescence. Cytometry 13 (1992) 48-59. 123. Marchetti D., Van N. T., Gametchu B., Thompson E. B., Kobayashi Y., Watanabe F. and Barlogie B.: Flow cytometric analysis of glucocorticoid receptor using monoclonal antibody and fluoresceinated ligand probes. Cancer Res. 49 (1989) 863-869. 124. Poulsen H. S., Jensen J. and Hermansen C.: Human breast cancer: heterogeneity of estrogen binding sites. Cancer 48 (1981) 1791-1793. 125. Bhisey A. N. and Freed J. J.: Ameboid movement induced in cultured macrophages by colchicine or vinblastine. Exp. Cell Res. 64 (1971) 419-429. 126. Akner G., Mossberg K., Sundqvist K.-G., Gustafsson J.-A. and Wikstr6m A.-C.: Evidence for reversible, non-microtubule and non-microfilament-dependent nuclear translocation of hsp90 after heat shock in human fibroblasts. Eur. J. Cell Biol. 58 (1992) 356-364. 127. Sanchez E. R.: Heat shock induces translocation to the nucleus

128. 129.

130.

131. 132.

133. 134.

135.

136.

137.

138.

139.

140.

of the unliganded glucocorticoid receptor. J. Biol. Chem. 267 (1992) 17-20. Szego C. M.: The lysosome as a mediator of hormone action. Recent Prog. Horm. Res. 30 (1974) 171-233. Hutchison K. A., Czar M. J. and Pratt W. B.: Evidence that the hormone-binding domain of the mouse glucocorticoid receptor directly represses DNA binding activity in a major portion of receptors that are "misfolded" after removal of hsp90. J. Biol. Chem. 267 (1992) 3190-3195. Martins V. R., Pratt W. B., Terracio L., Hirst M. A., Ringold G. M. and Housley P. R.: Demonstration by confocal microscopy that unliganded overexpressed glucocorticoid receptors are distributed in a nonrandom manner throughout all planes of the nucleus. Molec. Endocrinol. 5 (1991) 217-225. Liao S., Liang T. and Tymoczko J. L.: Ribonucleoprotein binding of steroid-"receptor" complexes. Nature New Biol. 241 (1973) 211-213. Verheijen R., Kuijpers H., Vooijs P., Van Venrooij W. and Rarnaekers F.: Distribution of the 70K U1 RNA-associated protein during interphase and mitosis. J. Cell Sci. 86 (1986) 173-190. Spector D. L.: Higher order nuclear organization: three-dimensional distribution of small nuclear ribonucleoprotein particles. Proc. Natn. Acad. Sci. U.S.A. 87 (1990) 147-151. Kr~rner A., Keller W., Appel B. and Lfihrmann R.: The 5' terminus of the RNA moiety of U1 snRNP particles is required for the splicing of messenger RNA precursors. Cell 28 (1984) 299-307. Towle A. C. and Sze P. Y.: Steroid binding to synaptic plasma membrane: differential binding of glucocorticoids and gonadal steroids, ft. Steroid Biochem. 18 (1983) 135-143. Liposits Z., Uht R. M., Harrison R. W., Gibbs F. P., Paull W. K. and Bohn M. C.: Ultrastructural localization of glucocorticoid receptor (GR) in hypothalamic paraventricular neurons synthesizing corticotropin releasing factor (CRF). Histochemistry 87 (1987) 407-412. Stalker A., Hermo L. and Antakly T.: Subcellular distribution of [3H]-dexarnethasone mesylate binding sites in Leydig cells using electron microscope radioautography. Am. J. Anat. 190 (1991) 19-30. Williams M. A. and Baba W. I.: The localization of [3H]aldosterone and [3H]cortisol within renal tubular cells by electron microscope autoradiography, ft. Endocrinol. 39 (1967) 543-554. Antakly T., Wang X. and Lefebvre Y.: The glucocorticoid receptor in the nuclear envelope: electron microscope immunogold localization and biochemical characterization. J. Cell. Biol. 109 (1989) 135a. Howell G. M., Gustafsson J.-A. and Lefebvre Y. A.: Glucocorticoid receptor identified on nuclear envelopes of male rat livers by affinity labeling and immunochemistry. Endocrinology 127 (1990) 1087-1096.